Research
A Section 508–conformant HTML version of this article is available at http://dx.doi.org/10.1289/ehp.1510308.
A Systems Biology Approach Reveals Converging Molecular Mechanisms that Link Different POPs to Common Metabolic Diseases Patricia Ruiz,1* Ally Perlina,2* Moiz Mumtaz,1 and Bruce A. Fowler 3 1Computational Toxicology and Methods Development Laboratory, Division of Toxicology and Human Health Sciences, Agency for Toxic Substances and Disease Registry, Atlanta, Georgia, USA; 2Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA; 3 Emory University Rollins School of Public Health, Atlanta, Georgia, USA
Background: A number of epidemiological studies have identified statistical associations between persistent organic pollutants (POPs) and metabolic diseases, but testable hypotheses regarding underlying molecular mechanisms to explain these linkages have not been published. Objectives: We assessed the underlying mechanisms of POPs that have been associated with metabolic diseases; three well-known POPs [2,3,7,8-tetrachlorodibenzodioxin (TCDD), 2,2´,4,4´,5,5´-hexachlorobiphenyl (PCB 153), and 4,4´-dichlorodiphenyldichloroethylene (p,p´-DDE)] were studied. We used advanced database search tools to delineate testable hypotheses and to guide laboratory-based research studies into underlying mechanisms by which this POP mixture could produce or exacerbate metabolic diseases. M ethods : For our searches, we used proprietary systems biology software (MetaCore™/ MetaDrug™) to conduct advanced search queries for the underlying interactions database, followed by directional network construction to identify common mechanisms for these POPs within two or fewer interaction steps downstream of their primary targets. These common downstream pathways belong to various cytokine and chemokine families with experimentally well-documented causal associations with type 2 diabetes. Conclusions: Our systems biology approach allowed identification of converging pathways leading to activation of common downstream targets. To our knowledge, this is the first study to propose an integrated global set of step-by-step molecular mechanisms for a combination of three common POPs using a systems biology approach, which may link POP exposure to diseases. Experimental evaluation of the proposed pathways may lead to development of predictive biomarkers of the effects of POPs, which could translate into disease prevention and effective clinical treatment strategies. Citation: Ruiz P, Perlina A, Mumtaz M, Fowler BA. 2016. A systems biology approach reveals converging molecular mechanisms that link different POPs to common metabolic diseases. Environ Health Perspect 124:1034–1041; http://dx.doi.org/10.1289/ehp.1510308
Introduction Persistent organic pollutants (POPs) are ubiquitous environmental contaminants. They include polychlorinated dibenzo-pdioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), polychlorinated biphenyls (PCBs), and organochlorine pesticides. 2,3,7,8-Tetrachlorodibenzodioxin (TCDD), a representative of the dioxin chemical family, is unintentionally produced during chlorine bleaching processes, drinking water chlorination, and incineration processes [Agency for Toxic Substances and Disease Registry (ATSDR) 2012]. 4,4´-Dichlorodiphenyl dichloroethylene (p,p´-DDE) is a metabolite of DDT that has been used as an insecticide for insect vectors of malaria and typhus (ATSDR 2008). Polychlorinated biphenyls (PCBs) are industrial chemicals principally used as heat exchange fluids in transformers and capacitors that were banned in the United States in 1977 (ATSDR 2011). Epidemiological studies have reported associations between POPs and metabolic diseases such as Type 2 diabetes (T2D), obesity, and metabolic syndrome, but the potential underlying mechanism(s) are not known (Langer et al. 2014; Lee et al. 2006,
1034
2010, 2011; Rylander et al. 2005; Suzuki et al. 2008). The three POPs evaluated in the present study (TCDD, PCB 153, and p,p´-DDE) have been associated with metabolic disorders in observational studies, but the potential molecular mechanisms that might underlie endocrine disruption and disease development are far from understood (Everett et al. 2007; Henriksen et al. 1997; Lee et al. 2014; Longnecker and Michalek 2000; Magliano et al. 2014; Rignell-Hydbom et al. 2007; Turyk et al. 2009; Wang et al. 2008). Because metabolic diseases are increasing in frequency throughout the world, further investigation and understanding of the possibility that exposure to POPs contributes to the etiology of diabetes, obesity, and cardiovascular disease is critical (Taylor et al. 2013; Thayer et al. 2012). Metabolic syndrome may affect up to 1 in 5 people, and its prevalence increases with age (Paoletti et al. 2006). It is estimated that ≤ 25% of the U.S. population has metabolic syndrome (Ford et al. 2004). Researchers have hypothesized that lowlevel POP exposure can cause metabolic changes through a network of pathways, including increased insulin resistance and obesity preceding the development of T2D volume
(Barouki et al. 2012; Barrett 2013; Lee et al. 2014; Taylor et al. 2013). Within this network, different POPs might also cause metabolic syndrome through slightly overlapping pathways to cause disturbances in glucose homeostasis. Such disturbances include inhibition of insulin action and induced down-regulation of master regulators of lipid homeostasis. The situation is further complicated by the realization that POP-induced alterations in epigenetic regulatory mechanisms may occur during sensitive developmental periods and lead to diseases such as obesity and T2D later in life (Barouki et al. 2012). In toxicology, systems biology facilitates the identification of important pathways and molecules from large data sets. These tasks can be extremely laborious when performed using a classical literature search. Computational systems biology offers more advantages than simply providing a high-throughput literature search engine; these tools may provide the basis for establishing hypotheses on potential links between environmental chemicals and human diseases. Comprehensive databases containing information on networks of human protein–protein interactions and protein– disease associations make it possible to identify such links. Experimentally determined target data for the specific chemical of interest can be *These authors contributed equally to this work. Current address for A. Perlina: Human Longevity, Inc., San Diego, California Address correspondence to P. Ruiz, Division of Toxicology and Human Health Sciences, Computational Toxicology and Methods Development Lab, Agency for Toxic Substances and Disease Registry, 1600 Clifton Rd., MS-F57, Atlanta, GA 30333. Telephone: (770) 488-3348. E-mail: [email protected] This study was presented at the 54th Society of Toxicology annual meeting (SOT2015). We are grateful to L.S. Birnbaum and D. Johnson for their valuable comments and advice during the preparation of this manuscript. We also thank D. Meadows for his editorial technical support. The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the Centers for Disease Control and Prevention/the Agency for Toxic Substances and Disease Registry. Mention of trade names is not an endorsement of any commercial product. The authors declare they have no actual or potential competing financial interests. Received: 5 June 2015; Accepted: 10 December 2015; Advance Publication: 18 December 2015; Final Publication: 1 July 2016.
124 | number 7 | July 2016 • Environmental Health Perspectives
POPs linkages to common metabolic diseases
uploaded and superimposed on these networks to obtain additional information that can be used to establish hypotheses on links between the chemical and human diseases. Such information can also be used to design rational animal- and cell-based laboratory experiments to test the established hypotheses. In this study, we examined potential linkages for combined exposures to three specific POPs, cellular pathway alterations, and metabolic disturbances related to the development of important clinical outcomes. We used an integrated global approach that brought together a) predictive chemical analyses based on compound structure and b) knowledge bases of chemogenomics data associating compounds with biological and toxicological properties. We then performed an in silico evaluation of the possible joint effects of POPs on metabolic pathways that could lead to metabolic diseases. We sought to discover common downstream activation targets for all three POPs as a mixture. Although inhibitory targets were also analyzed, we chose to focus on the genes that could ultimately be up-regulated and lead to increased abundance on the protein level. The rationale for this focus was to set the stage for the discovery of screening biomarkers, particularly those present in easily accessible tissues/fluids, the accessibility of which could be improved when increased in abundance, as opposed to depleted. It is hoped that these data will stimulate the formation of new, testable hypotheses to address some of the data gaps previously identified by Barrett (2013), La Merrill et al. (2013), and Taylor et al. (2013).
Methods Three POPs (p,p´-DDE, TCDD, and PCB 153) were selected for investigation in this study because they are commonly detected in the environment and in human tissues. Based upon data from the epidemiological and data mining literature noted above, they have also been linked with metabolic diseases such as T2D (Everett et al. 2007; Henriksen et al. 1997; Lee et al. 2010; Longnecker and Michalek 2000; Turyk et al. 2009; Wang et al. 2008). The majority of available POP studies have focused on these three POPs on an individual basis. To our knowledge, there are no published studies on their combined potential interactive effects at the molecular level in relation to clinical disease outcomes.
The Pathway Analysis Tools: Metacore™/Metadrug™ The molecular structure (.mol) files of three POPs [p,p´-DDE, TCDD, and PCB 153 (Figure 1)] were separately uploaded to MetaCore™/MetaDrug™, a proprietary systems biology software solution (Thomson
Reuters; originally developed by GeneGo, Inc.). This software is built on a proprietary database (MetaBase™) to allow functional and network analysis of primary and secondary effects of any query compound in the context of manually curated molecular interactions and pathways (Ekins et al. 2007). MetaCore™/MetaDrug™ are analytical tools built on top of a manually curated database of literature findings that support various types of molecular interactions and ontologies, including disease relationships. These tools help a user to analyze information from experimental results or to mine the underlying content from the MetaBase™ database content directly. Advanced Search is a java application tool in MetaCore™ that facilitates searching combined information, for example, “find all compounds that inhibit EGFR with IC50
A Section 508–conformant HTML version of this article is available at http://dx.doi.org/10.1289/ehp.1510308.
A Systems Biology Approach Reveals Converging Molecular Mechanisms that Link Different POPs to Common Metabolic Diseases Patricia Ruiz,1* Ally Perlina,2* Moiz Mumtaz,1 and Bruce A. Fowler 3 1Computational Toxicology and Methods Development Laboratory, Division of Toxicology and Human Health Sciences, Agency for Toxic Substances and Disease Registry, Atlanta, Georgia, USA; 2Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA; 3 Emory University Rollins School of Public Health, Atlanta, Georgia, USA
Background: A number of epidemiological studies have identified statistical associations between persistent organic pollutants (POPs) and metabolic diseases, but testable hypotheses regarding underlying molecular mechanisms to explain these linkages have not been published. Objectives: We assessed the underlying mechanisms of POPs that have been associated with metabolic diseases; three well-known POPs [2,3,7,8-tetrachlorodibenzodioxin (TCDD), 2,2´,4,4´,5,5´-hexachlorobiphenyl (PCB 153), and 4,4´-dichlorodiphenyldichloroethylene (p,p´-DDE)] were studied. We used advanced database search tools to delineate testable hypotheses and to guide laboratory-based research studies into underlying mechanisms by which this POP mixture could produce or exacerbate metabolic diseases. M ethods : For our searches, we used proprietary systems biology software (MetaCore™/ MetaDrug™) to conduct advanced search queries for the underlying interactions database, followed by directional network construction to identify common mechanisms for these POPs within two or fewer interaction steps downstream of their primary targets. These common downstream pathways belong to various cytokine and chemokine families with experimentally well-documented causal associations with type 2 diabetes. Conclusions: Our systems biology approach allowed identification of converging pathways leading to activation of common downstream targets. To our knowledge, this is the first study to propose an integrated global set of step-by-step molecular mechanisms for a combination of three common POPs using a systems biology approach, which may link POP exposure to diseases. Experimental evaluation of the proposed pathways may lead to development of predictive biomarkers of the effects of POPs, which could translate into disease prevention and effective clinical treatment strategies. Citation: Ruiz P, Perlina A, Mumtaz M, Fowler BA. 2016. A systems biology approach reveals converging molecular mechanisms that link different POPs to common metabolic diseases. Environ Health Perspect 124:1034–1041; http://dx.doi.org/10.1289/ehp.1510308
Introduction Persistent organic pollutants (POPs) are ubiquitous environmental contaminants. They include polychlorinated dibenzo-pdioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), polychlorinated biphenyls (PCBs), and organochlorine pesticides. 2,3,7,8-Tetrachlorodibenzodioxin (TCDD), a representative of the dioxin chemical family, is unintentionally produced during chlorine bleaching processes, drinking water chlorination, and incineration processes [Agency for Toxic Substances and Disease Registry (ATSDR) 2012]. 4,4´-Dichlorodiphenyl dichloroethylene (p,p´-DDE) is a metabolite of DDT that has been used as an insecticide for insect vectors of malaria and typhus (ATSDR 2008). Polychlorinated biphenyls (PCBs) are industrial chemicals principally used as heat exchange fluids in transformers and capacitors that were banned in the United States in 1977 (ATSDR 2011). Epidemiological studies have reported associations between POPs and metabolic diseases such as Type 2 diabetes (T2D), obesity, and metabolic syndrome, but the potential underlying mechanism(s) are not known (Langer et al. 2014; Lee et al. 2006,
1034
2010, 2011; Rylander et al. 2005; Suzuki et al. 2008). The three POPs evaluated in the present study (TCDD, PCB 153, and p,p´-DDE) have been associated with metabolic disorders in observational studies, but the potential molecular mechanisms that might underlie endocrine disruption and disease development are far from understood (Everett et al. 2007; Henriksen et al. 1997; Lee et al. 2014; Longnecker and Michalek 2000; Magliano et al. 2014; Rignell-Hydbom et al. 2007; Turyk et al. 2009; Wang et al. 2008). Because metabolic diseases are increasing in frequency throughout the world, further investigation and understanding of the possibility that exposure to POPs contributes to the etiology of diabetes, obesity, and cardiovascular disease is critical (Taylor et al. 2013; Thayer et al. 2012). Metabolic syndrome may affect up to 1 in 5 people, and its prevalence increases with age (Paoletti et al. 2006). It is estimated that ≤ 25% of the U.S. population has metabolic syndrome (Ford et al. 2004). Researchers have hypothesized that lowlevel POP exposure can cause metabolic changes through a network of pathways, including increased insulin resistance and obesity preceding the development of T2D volume
(Barouki et al. 2012; Barrett 2013; Lee et al. 2014; Taylor et al. 2013). Within this network, different POPs might also cause metabolic syndrome through slightly overlapping pathways to cause disturbances in glucose homeostasis. Such disturbances include inhibition of insulin action and induced down-regulation of master regulators of lipid homeostasis. The situation is further complicated by the realization that POP-induced alterations in epigenetic regulatory mechanisms may occur during sensitive developmental periods and lead to diseases such as obesity and T2D later in life (Barouki et al. 2012). In toxicology, systems biology facilitates the identification of important pathways and molecules from large data sets. These tasks can be extremely laborious when performed using a classical literature search. Computational systems biology offers more advantages than simply providing a high-throughput literature search engine; these tools may provide the basis for establishing hypotheses on potential links between environmental chemicals and human diseases. Comprehensive databases containing information on networks of human protein–protein interactions and protein– disease associations make it possible to identify such links. Experimentally determined target data for the specific chemical of interest can be *These authors contributed equally to this work. Current address for A. Perlina: Human Longevity, Inc., San Diego, California Address correspondence to P. Ruiz, Division of Toxicology and Human Health Sciences, Computational Toxicology and Methods Development Lab, Agency for Toxic Substances and Disease Registry, 1600 Clifton Rd., MS-F57, Atlanta, GA 30333. Telephone: (770) 488-3348. E-mail: [email protected] This study was presented at the 54th Society of Toxicology annual meeting (SOT2015). We are grateful to L.S. Birnbaum and D. Johnson for their valuable comments and advice during the preparation of this manuscript. We also thank D. Meadows for his editorial technical support. The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the Centers for Disease Control and Prevention/the Agency for Toxic Substances and Disease Registry. Mention of trade names is not an endorsement of any commercial product. The authors declare they have no actual or potential competing financial interests. Received: 5 June 2015; Accepted: 10 December 2015; Advance Publication: 18 December 2015; Final Publication: 1 July 2016.
124 | number 7 | July 2016 • Environmental Health Perspectives
POPs linkages to common metabolic diseases
uploaded and superimposed on these networks to obtain additional information that can be used to establish hypotheses on links between the chemical and human diseases. Such information can also be used to design rational animal- and cell-based laboratory experiments to test the established hypotheses. In this study, we examined potential linkages for combined exposures to three specific POPs, cellular pathway alterations, and metabolic disturbances related to the development of important clinical outcomes. We used an integrated global approach that brought together a) predictive chemical analyses based on compound structure and b) knowledge bases of chemogenomics data associating compounds with biological and toxicological properties. We then performed an in silico evaluation of the possible joint effects of POPs on metabolic pathways that could lead to metabolic diseases. We sought to discover common downstream activation targets for all three POPs as a mixture. Although inhibitory targets were also analyzed, we chose to focus on the genes that could ultimately be up-regulated and lead to increased abundance on the protein level. The rationale for this focus was to set the stage for the discovery of screening biomarkers, particularly those present in easily accessible tissues/fluids, the accessibility of which could be improved when increased in abundance, as opposed to depleted. It is hoped that these data will stimulate the formation of new, testable hypotheses to address some of the data gaps previously identified by Barrett (2013), La Merrill et al. (2013), and Taylor et al. (2013).
Methods Three POPs (p,p´-DDE, TCDD, and PCB 153) were selected for investigation in this study because they are commonly detected in the environment and in human tissues. Based upon data from the epidemiological and data mining literature noted above, they have also been linked with metabolic diseases such as T2D (Everett et al. 2007; Henriksen et al. 1997; Lee et al. 2010; Longnecker and Michalek 2000; Turyk et al. 2009; Wang et al. 2008). The majority of available POP studies have focused on these three POPs on an individual basis. To our knowledge, there are no published studies on their combined potential interactive effects at the molecular level in relation to clinical disease outcomes.
The Pathway Analysis Tools: Metacore™/Metadrug™ The molecular structure (.mol) files of three POPs [p,p´-DDE, TCDD, and PCB 153 (Figure 1)] were separately uploaded to MetaCore™/MetaDrug™, a proprietary systems biology software solution (Thomson
Reuters; originally developed by GeneGo, Inc.). This software is built on a proprietary database (MetaBase™) to allow functional and network analysis of primary and secondary effects of any query compound in the context of manually curated molecular interactions and pathways (Ekins et al. 2007). MetaCore™/MetaDrug™ are analytical tools built on top of a manually curated database of literature findings that support various types of molecular interactions and ontologies, including disease relationships. These tools help a user to analyze information from experimental results or to mine the underlying content from the MetaBase™ database content directly. Advanced Search is a java application tool in MetaCore™ that facilitates searching combined information, for example, “find all compounds that inhibit EGFR with IC50
